CN108445393A - A kind of permanent magnet synchronous motor fault detection method and system - Google Patents
A kind of permanent magnet synchronous motor fault detection method and system Download PDFInfo
- Publication number
- CN108445393A CN108445393A CN201810411549.8A CN201810411549A CN108445393A CN 108445393 A CN108445393 A CN 108445393A CN 201810411549 A CN201810411549 A CN 201810411549A CN 108445393 A CN108445393 A CN 108445393A
- Authority
- CN
- China
- Prior art keywords
- permanent magnet
- magnet synchronous
- synchronous motor
- image
- feature
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/34—Testing dynamo-electric machines
- G01R31/343—Testing dynamo-electric machines in operation
Abstract
The invention discloses a kind of permanent magnet synchronous motor fault detection method and system, disclosed method includes the following steps：Step S100：Acquire a series of N number of sequential raw image datas of normal and tape jam the thermal transformation of permanent magnet synchronous motor, j=1, j≤N；Step S200：Jth time raw image data is partitioned into the higher pixel region of permanent magnet synchronous motor surface temperature；Step S300：Jth time is subjected to feature extraction by the image data of image procossing；Step S400：Motor status classification is determined according to jth time extraction feature；Step S500：Judge j>Whether N is true, if not, j=j+1, return to step S200, on the contrary enter step S600；Step S600：Feature and its corresponding motor status classification are extracted by n times thermal transformation, establishes Diagnosing Faults of Electrical supporting vector machine model；Step S700：Fault distinguishing is carried out to collection in worksite feature with trained model.It can fully automatically, accurately judge electrical fault under continuous duty without being in direct contact motor.
Description
Technical field
The present invention relates to a kind of permanent magnet synchronous motor technical field more particularly to a kind of permanent magnet synchronous motor fault detect sides
Method and system.
Background technology
In recent years, with the fast development of modern science and technology, electromagnetic material especially rare-earth electromagnetic material property and work
Skill is gradually improved and improves, along with the high speed development of power electronics and power drives technology, automatic control technology, permanent magnetism
The performance of synchronous motor is become better and better.Furthermore permasyn morot has, and light weight, structure be simpler, small, characteristic
Well, the advantages that power density is big, many scientific research institutions, enterprise all actively develop the R＆D work of permanent magnet synchronous motor in effort,
Its application field will constantly expand.But the reasons such as the operating mode of usual motor is severe, serious vibration, operating ambient temperature are higher make
Motor is easy to break down, therefore permanent magnet synchronous motor Diagnosing Faults of Electrical is also an important field of research.
During Diagnosing Faults of Electrical, the feature of motor most common failure is most apparent from showing in the frequency of vibration signal
On the electric current of stator, although having reached certain reliability by frequency and failure of the current detection, both sides
Method needs to spend high detection device and a large amount of time, while while detecting can interrupt production.And due to being mostly at present
Differentiated that subjective factor is in the majority to be easy to cause erroneous judgement by artificial experience accumulation, lacks a kind of diagnosis of full-automatic science and sentence
Other method.
Therefore, how without in the case of being in direct contact motor, fully automatically, accurately judging under continuous duty
Permanent magnet synchronous motor failure, be those skilled in the art's urgent need to resolve the problem of.
Invention content
The object of the present invention is to provide a kind of permanent magnet synchronous motor fault detection method and system, can be not necessarily to directly connect
In the case of electric shock machine, the permanent magnet synchronous motor failure under continuous duty fully automatically, is accurately judged.
In order to solve the above technical problems, the present invention provides a kind of permanent magnet synchronous motor fault detection method, the method packet
Include following steps：
Step S100：A series of N number of sequential for acquiring normal and tape jam the thermal transformation of permanent magnet synchronous motor are former
Beginning image data, j=1, j≤N；
Step S200：A series of sequential raw image datas of the thermal transformation of jth time are used into image processing techniques
It is partitioned into the higher pixel region of permanent magnet synchronous motor surface temperature；
Step S300：A series of time sequence image datas of thermal transformation by the jth of image procossing time are carried out special
Sign extraction, it is straight including temperature variation curve feature, interframe variance vibration performance and the higher pixel region of motor surface temperature
Square figure correlated characteristic；
Step S400：Motor status classification is determined according to the feature of the thermal transformation of jth time extraction；
Step S500：Judge whether j ＞ N are true, if not, j=j+1, return to step S200 are entered step if setting up
S600；
Step S600：The feature and its corresponding motor status classification extracted by the thermal transformation of n times motor, are established
Permanent magnet synchronous motor fault diagnosis supporting vector machine model；
Step S700：Fault distinguishing is carried out to the permanent magnet synchronous motor at scene with trained model.
Preferably, raw image data is acquired using thermal infrared imager in step S100, and raw image data is gray scale
Image data.
Preferably, a series of sequential raw image datas of thermal transformation described in step S100 specifically refer to generally from
Cold conditions starts, and applies after specified step load permanent magnet synchronous motor temperature rise to establishing new thermal balance to permanent magnet synchronous motor
When a series of sequential original images, thermal infrared imager be set to prefixed time interval acquisition permanent magnet synchronous motor thermal transformation
A series of sequential raw image datas.
Preferably, it is specially in the step 200：
Step S210：Thermal infrared imager is collected to a series of sequential original-gray images of the thermal transformation of jth time
Be converted to true temperature pattern；
Step S220：A series of sequential temperature patterns of the thermal transformation of jth time are carried out Gaussian kernel to be filtered；
Step S230：It is random in a series of sequential temperature patterns of the thermal transformation of the jth time of filtered processing
The frame image for extracting hot steady state time section, calculates the threshold value of the extraction frame image segmentation, and threshold will be less than in the extraction frame image
The pixel assignment of value is 0, and other pixel values remain unchanged；
Step S240：It calculates and extracts frame image slices vegetarian refreshments mean μ described in step S2301And standard deviation sigma1, by the extraction frame
It is less than μ in image1-σ1Pixel be assigned a value of 0, other pixels remain unchanged, to be partitioned into permanent magnet synchronous motor surface temperature
Higher pixel region.
Preferably, it is specially in the step S300：
Step S310：Picture is determined according to a series of sequential temperature patterns of the thermal transformation of jth described in step S210 time
The curve of plain maximum temperature values changed over time；
Step S320：Regional edge is determined according to the higher pixel region of permanent magnet synchronous motor surface temperature described in step S240
Edge minimum value；
Step S330：Using the invariable pixel maximum temperature values of hot stable state and edges of regions minimum value as feature histogram
Section up and down, section is averagely divided into 10 equal portions, permanent magnet synchronous motor surface temperature is higher described in statistic procedure S240
Pixel region temperature histogram；
Step S340：Calculate mean value, the standard of the higher pixel region temperature histogram of permanent magnet synchronous motor surface temperature
Difference, skewness, kurtosis and entropy；
Step S350：The curvilinear characteristic of pixel maximum temperature values described in extraction step S310 changed over time；
Step S360：Calculate a series of sequential of the thermal transformation of the jth time of filtered processing described in step S220
The hot stable state interframe variance vibration performance of temperature pattern.
Preferably, the step S400 is specially：
Step S410：The feature of the extraction of the thermal transformation of jth time is formed into feature vector；
Step S420：Motor status classification is determined according to feature vector；
Preferably, the step S600 is specially：
Step S610：The feature vector and its corresponding motor status classification extracted by the thermal transformation of n times motor
Establish the training set of grader；
Step S620：Grader uses support vector machines, kernel function to use Radial basis kernel function, and the training set of classification is to carry
The feature vector taken is as input, and using motor status classification as output, Training Support Vector Machines obtain permanent magnet synchronous motor event
Barrier diagnosis supporting vector machine model.
The present invention also provides a kind of permanent magnet synchronous motor fault detection systems, including image capture module, image to locate in advance
Module, characteristic extracting module, model building module and fault detection module are managed, wherein：
Image capture module, a N number of system for normal and tape jam the thermal transformation for acquiring permanent magnet synchronous motor
Row sequential raw image data, j=1, j≤N are sent to image pre-processing module；
Image pre-processing module, a series of sequential of the thermal transformation of the jth time for sending image capture module
Raw image data is partitioned into the higher pixel region of permanent magnet synchronous motor surface temperature using image processing techniques, is sent to spy
Levy extraction module；
Characteristic extracting module, the thermal change mistake of the jth by image procossing time for sending image pre-processing module
A series of time sequence image datas of journey extract feature, including temperature variation curve feature, interframe variance vibration performance
With the higher pixel region histogram correlated characteristic of motor surface temperature, it is sent to model building module；
Model building module, the extraction feature for being sent according to characteristic extracting module determine motor status classification；Judge j
>Whether N is true, if not, j=j+1 returns to image pre-processing module, if so, then pass through the thermal change mistake of n times motor
Journey feature extraction and motor status classification determine, establish permanent magnet synchronous motor fault diagnosis supporting vector machine model, are sent to event
Hinder detection module；
Fault detection module, for carrying out failure using the permanent magnet synchronous motor of the trained model pair of model building module
Differentiate.
Preferably, in described image acquisition module raw image data is acquired using thermal infrared imager.
Preferably, when the thermal infrared imager acquisition raw image data, thermal infrared imager puts permanent-magnet synchronous to be aligned
Motor center side, thermal infrared imager and permanent magnet synchronous motor are located at same level height, the choosing of horizontal distance between the two
It includes entire permanent magnet synchronous motor to take thermal infrared imager the image collected to be made, and both acquisitions relative position is fixed not every time
Become.
Can fully automatically, accurately judge to include stator under continuous duty without being in direct contact motor
The permanent magnet synchronous motors failures such as turn-to-turn fault, bearing fault, heat dissipation failure and demagnetization failure.
Description of the drawings
Fig. 1 is a kind of flow chart for permanent magnet synchronous motor fault detection method that the first embodiment provides；
Fig. 2 is a kind of flow chart for permanent magnet synchronous motor fault detection method that second of embodiment provides；
Fig. 3 is normal motor and faulty motor maximum temperature change curve；
Fig. 4 is a kind of structure diagram of permanent magnet synchronous motor fault detection system provided by the invention；
Fig. 5 is the schematic front view of permanent magnet synchronous motor and thermal infrared imager provided by the invention placement position；
Fig. 6 is the schematic top plan view of permanent magnet synchronous motor and thermal infrared imager provided by the invention placement position.
Specific implementation mode
In order that those skilled in the art will better understand the technical solution of the present invention, below in conjunction with the accompanying drawings to the present invention
It is described in further detail.
Referring to Fig. 1, Fig. 1 is a kind of flow for permanent magnet synchronous motor fault detection method that the first embodiment provides
Figure.
A kind of permanent magnet synchronous motor fault detection method, the described method comprises the following steps：
Step S100：A series of N number of sequential for acquiring normal and tape jam the thermal transformation of permanent magnet synchronous motor are former
Beginning image data, j=1, j≤N, N are according to actual needs and empirically determined；
Step S200：A series of sequential raw image datas of the thermal transformation of jth time are used into image processing techniques
It is partitioned into the higher pixel region of permanent magnet synchronous motor surface temperature；
Step S300：A series of time sequence image datas of thermal transformation by the jth of image procossing time are carried out special
Sign extraction, it is straight including temperature variation curve feature, interframe variance vibration performance and the higher pixel region of motor surface temperature
Square figure correlated characteristic；
Step S400：Motor status classification is determined according to the feature of the thermal transformation of jth time extraction；
Step S500：Judge j>Whether N is true, if not, j=j+1, return to step S200 are entered step if setting up
S600；
Step S600：The feature and its corresponding motor status classification extracted by the thermal transformation of n times motor, are established
Permanent magnet synchronous motor fault diagnosis supporting vector machine model；
Step S700：Fault distinguishing is carried out to the permanent magnet synchronous motor at scene with trained model.
The one of the thermal transformation of normal permanent magnet synchronous motor and the permanent magnet synchronous motor of several tape jams is acquired respectively
Serial sequential raw image data.Preferably, acquired respectively using thermal infrared imager include permanent magnet synchronous motor normal work
And the permanent magnet synchronous motor thermal transformation of the various failures such as interturn in stator windings failure, bearing fault, fault of eccentricity, demagnetization failure
A series of sequential original-gray image data.A series of sequential raw image datas of thermal transformation specifically refer to generally from
Cold conditions starts, and applies after specified step load permanent magnet synchronous motor temperature rise to establishing new thermal balance to permanent magnet synchronous motor
When a series of sequential original images.Thermal infrared imager is set to prefixed time interval acquisition permanent magnet synchronous motor thermal transformation
A series of sequential raw image datas.The prefixed time interval is rule of thumb set.
A series of sequential raw image datas of certain primary thermal transformation are partitioned into forever using image processing techniques
The higher pixel region of magnetic-synchro motor surface temperature.Will by image procossing this time a series of time sequence image data into
Row feature extraction.Extraction is higher including temperature variation curve feature, interframe variance vibration performance and motor surface temperature
Pixel region histogram correlated characteristic.The feature extracted according to the thermal transformation of this time determines motor status classification.According to step
The method of S200 to step S400 collects normal and different faults type permanent magnet synchronous motor thermal transformation by all
A series of sequential raw image datas carry out image procossing and feature extraction, and motor status class is determined according to the feature of extraction
Not.The feature and its corresponding motor status classification extracted by all thermal transformations, establish permanent magnet synchronous motor failure
Diagnose supporting vector machine model.Fault distinguishing is carried out to the permanent magnet synchronous motor of collection in worksite using trained model, is judged
The operating status of motor then judges which kind of permanent magnet synchronous motor failure belonged to, and alarm if there is failure.
In further scheme, it can acquire that model of the same race is normal and the permanent magnet synchronous motor of different faults type
A series of sequential raw image datas of thermal transformation, and then by image procossing, feature extraction, and it is true according to extraction feature
Determine to establish after motor status classification the fault diagnosis supporting vector machine model of the permanent magnet synchronous motor of the type.To different model
Permanent magnet synchronous motor establishes the fault diagnosis supporting vector machine model of its corresponding permanent magnet synchronous motor respectively.For different type
Permanent magnet synchronous motor carry out breakdown judge when can pre-enter motor model after, then use the corresponding permanent-magnet synchronous of the model
The fault diagnosis supporting vector machine model of motor carries out fault distinguishing to the permanent magnet synchronous motor.
Fully automatically, it can accurately judge continuous duty without being in direct contact motor using this method
Include the permanent magnet synchronous motors failures such as interturn in stator windings failure, bearing fault, heat dissipation failure and demagnetization failure down.
Referring to Fig. 2 to Fig. 3, Fig. 2 is a kind of permanent magnet synchronous motor fault detection method that second of embodiment provides
Flow chart, Fig. 3 are normal motor and faulty motor maximum temperature change curve.
A kind of permanent magnet synchronous motor fault detection method, the described method comprises the following steps：
Step S100：Thermal infrared imager acquires N number of the one of normal and tape jam the thermal transformation of permanent magnet synchronous motor
Serial sequential original-gray image data, j=1, j≤N.
Step S210：Thermal infrared imager is collected to a series of sequential original-gray images of the thermal transformation of jth time
Be converted to true temperature pattern.
Since the original image of thermal infrared imager acquisition is gray level image, the value of each pixel of gray level image does not represent
True temperature need to be converted to obtain true temperature pattern according to following formula (1)：
Wherein, T (x, y) is the temperature value at image (x, y), Wtot(x, y) is the image (x, y) of thermal infrared imager acquisition
The global radiation at place, τatmIt is the transmitance of room air, σ is Stefan-Boltzmann constant, TreflIt is reflected temperature, TatmIt is
Indoor temperature, εobjIt is the emissivity of object.
Step S220：A series of sequential temperature patterns of the thermal transformation of jth time are carried out Gaussian kernel to be filtered, are used
3 × 3 Gaussian kernels remove a series of noise in sequential temperature patterns, wherein the formula filtered is as follows：
Wherein I (x, y) is the pixel value at temperature pattern (x, y) after gaussian filtering, and Kernel (k, l) is 3 × 3 Gaussian kernels
Pixel value of the function at (k, l).
Step S230：It is random in a series of sequential temperature patterns of the thermal transformation of the jth time of filtered processing
The frame for extracting hot steady state time section calculates T with maximum variance between clusterstresholdAs the threshold value of image segmentation, by the extraction
The pixel assignment for being less than threshold value in frame image is 0, and other pixel values remain unchanged.
Step S240：It calculates and extracts frame image slices vegetarian refreshments mean μ described in step S2301And standard deviation sigma1, by the extraction frame
It is less than μ in image1-σ1Pixel be assigned a value of 0, other pixels remain unchanged, to be partitioned into permanent magnet synchronous motor surface temperature
Higher pixel region.Mean μ1And standard deviation sigma1Obtained by following formula：
Wherein, I (c) indicates the temperature value of c-th of pixel, adds up to n pixel.
Step S310：Picture is determined according to a series of sequential temperature patterns of the thermal transformation of jth described in step S210 time
Plain maximum temperature values TmaxThe curve T changed over timemax(t), as shown in figure 3, be same type normal motor and with therefore
Hinder motor TmaxThe temperature curve T changed over timemax(t) and the corresponding temperature rise time integral region.And motor temperature
The area in the region of rise time integral can be used as one of feature of fault diagnosis and be input in support vector machines.
Step S320：Regional edge is determined according to the higher pixel region of permanent magnet synchronous motor surface temperature described in step S240
Edge minimum value Tmin；
Step S330：By the invariable pixel maximum temperature values T of hot stable state in step S310maxIt is determined with step S320
Edges of regions minimum value TminAs the section up and down of feature histogram, section is averagely divided into 10 equal portions, statistic procedure S240
The higher pixel region temperature histogram of permanent magnet synchronous motor surface temperature, it is specific as follows：
Wherein, S (xi) it is xiThe number of pixels of temperature range；H(xi) it is xiThe frequency that temperature range temperature occurs.
Step S340：Calculate the higher pixel region temperature histogram of permanent magnet synchronous motor surface temperature that step 330 obtains
The mean μ of figure2, standard deviation sigma2, skewness Sk, kurtosis K and entropy E：
Step S350：Pixel maximum temperature values T described in extraction step S310maxThe curve T changed over timemax(t) special
Sign：
Wherein, Area is the integral of ascending temperature in the rise time, tsRefer to reaching new hot stable state 10% to rise to
Rise time section needed for 90%, Tmax(t) be motor t moment temperature maximum value；TatmIt is indoor environment temperature.
Step S360：Calculate a series of sequential of the thermal transformation of the jth time of filtered processing described in step S220
The hot stable state interframe variance vibration performance V of temperature pattern：
Wherein, V is the maximum value in the interframe variance of hot stable state sequential temperature pattern, It(z) it is the of hot stable state t frames
The temperature value of z pixel, a frame image add up to n pixel value；It is the temperature that hot stable state amounts to f frames z-th of pixel of image
It is worth average value, a series of sequential temperature patterns are f frames.
Step S410：The feature of the extraction of the thermal transformation of jth time is formed into feature vector, thermal transformation
A series of sequential temperature patterns can obtain corresponding 19 dimensional feature vector X=[Area, V, Tmax, Tmin, μ2, σ2, Sk, E, K, H
(x1) ... H (xi) ... H (x10), wherein H (xi) it is most hot-zone xthiThe frequency of temperature range, TmaxIt is the constant of hot stable state
Value.
Step S420：Motor status classification Y is determined according to feature vector, X.The determination of classification Y is by experiment electric motor status categories
It determines, the fault type of such as detectable permanent magnet synchronous motor has interturn in stator windings failure Y1, bearing fault Y2, heat dissipation failure Y3, move back
Magnetic failure Y4, the also classification Y of normal motor0。
Step S500：Judge whether j ＞ N are true, if not, j=j+1, return to step S200 are entered step if setting up
S610；
Step S610：The feature vector and its corresponding motor status classification extracted by the thermal transformation of n times motor
Establish the training set of grader；
Step S620：Grader uses support vector machines, kernel function to use Radial basis kernel function, and the training set of classification is to carry
The feature vector, X taken is as input, and using motor status classification Y as output, Training Support Vector Machines obtain permanent magnet synchronous motor
Fault diagnosis supporting vector machine model.
Step S700：Fault distinguishing is carried out to live permanent magnet synchronous motor with trained model.Thermal infrared imager acquires
Then a series of live sequential original-gray image data of permanent magnet synchronous motor use step S210 to step S360 to obtain scene
The feature of permanent magnet synchronous motor when field failure judges, judges electricity with trained model according to the feature of permanent magnet synchronous motor
The operating status of machine then judges which kind of magneto failure belonged to, and alarm if there is failure.
Referring to fig. 4 to fig. 6, Fig. 4 is a kind of structure diagram of permanent magnet synchronous motor fault detection system provided by the invention,
Fig. 5 is the schematic front view of permanent magnet synchronous motor and thermal infrared imager provided by the invention placement position, and Fig. 6 is offer of the present invention
Permanent magnet synchronous motor and thermal infrared imager placement position schematic top plan view.
The present invention also provides a kind of permanent magnet synchronous motor fault detection systems, including image capture module 1, and image is located in advance
Module 2, characteristic extracting module 3, model building module 4 and fault detection module 5 are managed, wherein：
Image capture module 1, the permanent magnet synchronous motor thermal change for acquiring normal permanent magnet synchronous motor and tape jam
A series of N number of sequential raw image datas of process, j=1, j≤N, N is according to actual needs and empirically determined, is sent to image
Preprocessing module 2；
Image pre-processing module 2, the thermal transformation of jth for sending image capture module 1 time it is a series of when
Sequence raw image data is partitioned into the higher pixel region of permanent magnet synchronous motor surface temperature using image processing techniques, is sent to
Characteristic extracting module 3；
Characteristic extracting module 3, the thermal change of the jth by image procossing time for sending image pre-processing module 2
A series of time sequence image datas of process extract feature, are vibrated including temperature variation curve feature, interframe variance special
The higher pixel region histogram correlated characteristic of motor surface temperature of seeking peace, is sent to model building module 4；
Model building module 4, the extraction feature for being sent according to characteristic extracting module 3 determine motor status classification；Sentence
Disconnected j>Whether N is true, if not, j=j+1 returns to image pre-processing module, if so, then pass through the thermal change of n times motor
Process feature extracts and motor status classification determines, establishes permanent magnet synchronous motor fault diagnosis supporting vector machine model, is sent to
Fault detection module 5；
Fault detection module 5, for being carried out to live permanent magnet synchronous motor using 4 trained model of model building module
Fault distinguishing.
Image capture module 1 acquires the one of normal and several tape jams the thermal transformation of permanent magnet synchronous motor respectively
Serial sequential raw image data.Preferably, using thermal infrared imager acquisition include permanent magnet synchronous motor normal work and
The one of the permanent magnet synchronous motor thermal transformations of various failures such as interturn in stator windings failure, bearing fault, fault of eccentricity, demagnetization failure
Serial sequential raw image data.A series of sequential raw image datas of thermal transformation specifically refer to generally open from cold conditions
Begin, permanent magnet synchronous motor temperature rise after specified step load is applied to one when establishing new thermal balance to permanent magnet synchronous motor
Serial sequential original image.Thermal infrared imager is set to a system of prefixed time interval acquisition permanent magnet synchronous motor thermal transformation
Row sequential raw image data.
Image pre-processing module 2 is by a series of sequential raw image datas of certain primary thermal transformation using at image
Reason technology is partitioned into the higher pixel region of permanent magnet synchronous motor surface temperature.Characteristic extracting module 3 will be by image procossing
The a series of time sequence image data of this time carries out feature extraction.Extraction is including temperature variation curve feature, interframe variance
Vibration performance and the higher pixel region histogram correlated characteristic of motor surface temperature.Model building module 4 is according to the thermal change of this time
The feature for changing procedure extraction determines motor status classification.According to the method for step images preprocessing module 2 and characteristic extracting module 3
A series of all sequential raw image datas for collecting different faults type permanent magnet synchronous motor thermal transformation are subjected to figure
As processing and feature extraction, and motor status classification is determined according to the feature of extraction.It is extracted by all thermal transformations
Feature and its corresponding motor status classification, establish permanent magnet synchronous motor fault diagnosis supporting vector machine model.Fault detect mould
Block 5 carries out fault distinguishing using trained model to live permanent magnet synchronous motor, judges the operating status of motor, if there is
Failure then judges which kind of permanent magnet synchronous motor failure belonged to, and alarms.
In further scheme, model different faults type permanent magnet synchronous motor thermal transformation of the same race can be acquired
A series of sequential raw image datas, and then by image preprocessing, feature extraction, and motor shape is determined according to extraction feature
The fault diagnosis supporting vector machine model of the permanent magnet synchronous motor of the type is established after state classification.To the permanent-magnet synchronous of different model
After motor establishes the fault diagnosis supporting vector machine model of its corresponding permanent magnet synchronous motor respectively, for different types of permanent magnetism
Synchronous motor carries out after can pre-entering motor model when breakdown judge, then uses the corresponding permanent magnet synchronous motor of the model
Fault diagnosis supporting vector machine model to the permanent magnet synchronous motor carry out fault distinguishing.
Can fully automatically, accurately judge to include stator under continuous duty without being in direct contact motor
The permanent magnet synchronous motors failures such as turn-to-turn fault, bearing fault, heat dissipation failure and demagnetization failure.
As shown in Figure 5 and Figure 6, when the thermal infrared imager acquisition raw image data, in tested permanent magnet synchronous motor mesh
A thermal infrared imager is installed in mark front, and thermal infrared imager has timed capture thermal image and is transferred to the work(of computer center
Can, for temperature-measuring range at -20 DEG C -300 DEG C, temperature resolution is less than 0.1 DEG C.Thermal infrared imager puts permanent magnet synchronous motor to be aligned
Center side, thermal infrared imager and permanent magnet synchronous motor are located at same level height, and the selection of horizontal distance between the two is wanted
It includes entire permanent magnet synchronous motor to make thermal infrared imager the image collected, and both acquisitions relative position immobilizes every time.Number
Electromagnetic shielding is carried out according to transmission cable, optimizing the length of transmission cable reduces signal attenuation, adjusts the vision of thermal infrared imager
Sensor aperture and focal length ensure that collected heat distribution image clearly is accurate.
A kind of permanent magnet synchronous motor fault detection method provided by the present invention and system are described in detail above.
Principle and implementation of the present invention are described for specific case used herein, and the explanation of above example is only used
Understand core of the invention thought in help.It should be pointed out that for those skilled in the art, not departing from
, can be with several improvements and modifications are made to the present invention under the premise of the principle of the invention, these improvement and modification also fall into this hair
In bright scope of the claims.
Claims (10)
1. a kind of permanent magnet synchronous motor fault detection method, which is characterized in that the described method comprises the following steps：
Step S100：Acquire a series of N number of sequential original graphs of normal and tape jam the thermal transformation of permanent magnet synchronous motor
As data, j=1, j≤N；
Step S200：A series of sequential raw image datas of the thermal transformation of jth time are divided using image processing techniques
Go out the higher pixel region of permanent magnet synchronous motor surface temperature；
Step S300：A series of time sequence image datas of thermal transformation by the jth of image procossing time are carried out feature to carry
It takes, including temperature variation curve feature, interframe variance vibration performance and the higher pixel region histogram of motor surface temperature
Correlated characteristic；
Step S400：Motor status classification is determined according to the feature of the thermal transformation of jth time extraction；
Step S500：Judge j>Whether N is true, if not, j=j+1, return to step S200 enter step S600 if setting up；
Step S600：The feature and its corresponding motor status classification extracted by the thermal transformation of n times motor, establish permanent magnetism
Synchronous motor fault diagnosis supporting vector machine model；
Step S700：Fault distinguishing is carried out to the permanent magnet synchronous motor at scene with trained model.
2. permanent magnet synchronous motor fault detection method according to claim 1, which is characterized in that using red in step S100
Outer thermal imaging system acquires raw image data, and raw image data is greyscale image data.
3. permanent magnet synchronous motor fault detection method according to claim 2, which is characterized in that hot described in step S100
A series of sequential raw image datas of change procedure specifically refer to generally since cold conditions, apply to permanent magnet synchronous motor specified
Permanent magnet synchronous motor temperature rise is to a series of sequential original images when establishing new thermal balance, infrared thermal imagery after step load
Instrument is set to a series of sequential raw image datas of prefixed time interval acquisition permanent magnet synchronous motor thermal transformation.
4. permanent magnet synchronous motor fault detection method according to claim 3, which is characterized in that in the step 200
Specially：
Step S210：Thermal infrared imager is collected to a series of sequential original-gray images conversion of the thermal transformation of jth time
For true temperature pattern；
Step S220：A series of sequential temperature patterns of the thermal transformation of jth time are carried out Gaussian kernel to be filtered；
Step S230：It is randomly selected in a series of sequential temperature patterns of the thermal transformation of the jth time of filtered processing
One frame image of hot steady state time section, calculates the threshold value of the extraction frame image segmentation, and threshold value will be less than in the extraction frame image
Pixel assignment is 0, and other pixel values remain unchanged；
Step S240：It calculates and extracts frame image slices vegetarian refreshments mean μ described in step S2301And standard deviation sigma1, by the extraction frame image
In be less than μ1-σ1Pixel be assigned a value of 0, other pixels remain unchanged, higher to be partitioned into permanent magnet synchronous motor surface temperature
Pixel region.
5. permanent magnet synchronous motor fault detection method according to claim 4, which is characterized in that have in the step S300
Body is：
Step S310：Determine pixel most according to a series of sequential temperature patterns of the thermal transformation of jth described in step S210 time
The curve of big temperature value changed over time；
Step S320：Determine edges of regions most according to the higher pixel region of permanent magnet synchronous motor surface temperature described in step S240
Small value；
Step S330：Using the invariable pixel maximum temperature values of hot stable state and edges of regions minimum value as the upper of feature histogram
Section, is averagely divided into 10 equal portions by lower section, the higher pixel of permanent magnet synchronous motor surface temperature described in statistic procedure S240
Regional temperature histogram；
Step S340：Calculate the mean value of the higher pixel region temperature histogram of permanent magnet synchronous motor surface temperature, standard deviation, partially
State, kurtosis and entropy；
Step S350：The curvilinear characteristic of pixel maximum temperature values described in extraction step S310 changed over time；
Step S360：Calculate a series of sequential temperature of the thermal transformation of the jth time of filtered processing described in step S220
The hot stable state interframe variance vibration performance of image.
6. permanent magnet synchronous motor fault detection method according to claim 5, which is characterized in that the step S400 is specific
For：
Step S410：The feature of the extraction of the thermal transformation of jth time is formed into feature vector；
Step S420：Motor status classification is determined according to feature vector.
7. permanent magnet synchronous motor fault detection method according to claim 6, which is characterized in that the step S600 is specific
For：
Step S610：The feature vector and its corresponding motor status classification extracted by the thermal transformation of n times motor are established
The training set of grader；
Step S620：Grader uses support vector machines, kernel function uses Radial basis kernel function, and the training set of classification is to extract
Feature vector is as input, and using motor status classification as output, Training Support Vector Machines obtain permanent magnet synchronous motor failure and examine
Disconnected supporting vector machine model.
8. a kind of permanent magnet synchronous motor fault detection system, which is characterized in that including image capture module, image preprocessing mould
Block, characteristic extracting module, model building module and fault detection module, wherein：
Image capture module, for acquire permanent magnet synchronous motor normal and tape jam thermal transformation it is N number of a series of when
Sequence raw image data, j=1, j≤N are sent to image pre-processing module；
A series of sequential of image pre-processing module, the thermal transformation of the jth time for sending image capture module are original
Image data is partitioned into the higher pixel region of permanent magnet synchronous motor surface temperature using image processing techniques, is sent to feature and carries
Modulus block；
Characteristic extracting module, the thermal transformation of the jth by image procossing time for sending image pre-processing module
A series of time sequence image datas extract feature, including temperature variation curve feature, interframe variance vibration performance and electricity
The higher pixel region histogram correlated characteristic of machine surface temperature, is sent to model building module；
Model building module, the extraction feature for being sent according to characteristic extracting module determine motor status classification；Judge j>N is
No establishment, if not, j=j+1 returns to image pre-processing module, if so, the thermal transformation for then passing through n times motor is special
Sign extraction and motor status classification determine, establish permanent magnet synchronous motor fault diagnosis supporting vector machine model, are sent to failure inspection
Survey module；
Fault detection module, for carrying out failure to the permanent magnet synchronous motor at scene using the trained model of model building module
Differentiate.
9. permanent magnet synchronous motor fault detection system according to claim 8, which is characterized in that described image acquisition module
It is middle to acquire raw image data using thermal infrared imager.
10. permanent magnet synchronous motor fault detection system according to claim 9, which is characterized in that the thermal infrared imager
When acquiring raw image data, thermal infrared imager puts permanent magnet synchronous motor center side to be aligned, thermal infrared imager and permanent magnetism
Synchronous motor is located at same level height, and the selection of horizontal distance between the two will make thermal infrared imager the image collected packet
Entire permanent magnet synchronous motor is included, both acquisitions relative position immobilizes every time.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810411549.8A CN108445393B (en) | 2018-05-02 | 2018-05-02 | A kind of permanent magnet synchronous motor fault detection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810411549.8A CN108445393B (en) | 2018-05-02 | 2018-05-02 | A kind of permanent magnet synchronous motor fault detection method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108445393A true CN108445393A (en) | 2018-08-24 |
| CN108445393B CN108445393B (en) | 2019-10-29 |
Family
ID=63202642
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810411549.8A Active CN108445393B (en) | 2018-05-02 | 2018-05-02 | A kind of permanent magnet synchronous motor fault detection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108445393B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114115175A (en) * | 2021-10-29 | 2022-03-01 | 江苏大学 | High-impedance connection fault diagnosis system of permanent magnet synchronous motor control system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101163364A (en) * | 2006-10-13 | 2008-04-16 | 北方工业大学 | Monitoring system and method for monitoring operation of electric element using the system |
| CN107247230A (en) * | 2017-07-03 | 2017-10-13 | 浙江大学 | A kind of electric rotating machine state monitoring method based on SVMs and data-driven |
| CN107609574A (en) * | 2017-08-18 | 2018-01-19 | 上海电力学院 | Wind turbines fault early warning method based on data mining |
| WO2018020545A1 (en) * | 2016-07-25 | 2018-02-01 | 三菱電機株式会社 | Electric motor diagnosis device |
-
2018
- 2018-05-02 CN CN201810411549.8A patent/CN108445393B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101163364A (en) * | 2006-10-13 | 2008-04-16 | 北方工业大学 | Monitoring system and method for monitoring operation of electric element using the system |
| WO2018020545A1 (en) * | 2016-07-25 | 2018-02-01 | 三菱電機株式会社 | Electric motor diagnosis device |
| CN107247230A (en) * | 2017-07-03 | 2017-10-13 | 浙江大学 | A kind of electric rotating machine state monitoring method based on SVMs and data-driven |
| CN107609574A (en) * | 2017-08-18 | 2018-01-19 | 上海电力学院 | Wind turbines fault early warning method based on data mining |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114115175A (en) * | 2021-10-29 | 2022-03-01 | 江苏大学 | High-impedance connection fault diagnosis system of permanent magnet synchronous motor control system |
| CN114115175B (en) * | 2021-10-29 | 2024-01-05 | 江苏大学 | High-impedance connection fault diagnosis system of permanent magnet synchronous motor control system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108445393B (en) | 2019-10-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105158257B (en) | Slide plate measurement method and device | |
| EP2851841A2 (en) | System and method of alerting a driver that visual perception of pedestrian may be difficult | |
| KR101747216B1 (en) | Apparatus and method for extracting target, and the recording media storing the program for performing the said method | |
| KR101759496B1 (en) | System and Method for Classification of PCB fault and Type of Fault | |
| CN105631418A (en) | People counting method and device | |
| CN104102069B (en) | A kind of focusing method of imaging system and device, imaging system | |
| KR20080022189A (en) | Object detection device, and object detection device for elevator | |
| CN102930287A (en) | Overlook-based detection and counting system and method for pedestrians | |
| JP2010003177A (en) | Image processor | |
| CN101569194A (en) | Network surveillance system | |
| CN107238796A (en) | A kind of maintainability test equipment for wiring board | |
| CN103552890A (en) | System and method for detecting number of people in lift car | |
| CN108445393B (en) | A kind of permanent magnet synchronous motor fault detection method and system | |
| CN113781537A (en) | Track elastic strip fastener defect identification method and device and computer equipment | |
| CN105915785A (en) | Double-camera shadedness determining method and device, and terminal | |
| CN111368698A (en) | Subject recognition method, subject recognition device, electronic device, and medium | |
| CN113188509B (en) | Distance measurement method and device, electronic equipment and storage medium | |
| JP2010140315A (en) | Object detection device | |
| JP5253195B2 (en) | Object detection device | |
| CN102262782A (en) | Digital image evidence obtaining method by utilizing CFA (color filter array) resampling interpolation and splicing positioning | |
| WO2014058165A1 (en) | Image monitoring apparatus for estimating size of singleton, and method therefor | |
| CN105427303B (en) | A kind of vision measurement and method of estimation of substation's legacy | |
| CN107495978A (en) | X-ray shooting system and image-pickup method | |
| CN110633680A (en) | Video-based people number abnormity detection method | |
| CN104376742B (en) | A kind of parking space state detection method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |